Abstract Children with substantial African ancestry have long been known to have half or less the rate of B-cell acute lymphoblastic leukemia (B-ALL) than do children with other continental ancestries. This is true both in international comparisons of rates of ALL in African nations to those elsewhere, and in comparing rate of B-ALL in African-American (AA) children to that in European-American (EA) children in the United States. The inverse association of African ancestry with incidence of B-ALL is independent of established perinatal risk factors for the disease. Moreover, AA children have lower incidence despite having greater exposure to many putatively causal environmental risk factors for B-ALL than do EA children. Common genetic variants established by genomewide association studies incompletely explain the deficit of B-ALL in AA children, suggesting undiscovered contributing genetic factors may be detected by admixture mapping. We have assembled existing DNA samples and data for 930 B-ALL patients with AA ancestry and will additionally accrue ~590 over the life of the project. We will conduct admixture mapping in the assembled group of patients to detect new genetic loci and new variants at established loci associated with occurrence of B-ALL. In addition, we will examine admixture in association with clinical characteristics at diagnosis and survival. Candidate genes/variants will be functionally evaluated through both in silico and in vitro techniques. The proposed research will potentially answer a long- standing mystery by revealing critical genes or loci that explain the comparative deficit of B-ALL in AA compared to EA children. In addition, we may uncover genes or variants associated with the worse characteristics at presentation in AA patients as well as with worse survival, which will indicate avenues for improving outcome among AA children.